Combined minimally invasive surgery for a dumbbell tumor using spinal endoscopy and robot-assisted intrathoracic surgery: a case report

Abstract

Background: The resection of dumbbell tumors requires careful planning that combines neurosurgical and thoracic procedures, as intra-thoracic manipulation alone may result in uncontrolled bleeding in the spinal cord owing to tumor traction and spinal cord injury. In recent years, many studies have reported intra-thoracic manipulation using robot-assisted surgery for improved safety and minimal invasiveness. However, several neurosurgical procedures of hybrid surgery have been performed using laminectomy. Herein, we describe a less invasive approach that combines robot-assisted surgery with spinal endoscopy to successfully resect a dumbbell tumor. Case Description: A 77-year-old man presented with a tumor in the posterior mediastinum. Computed tomography showed a well-demarcated, 3-cm sized dumbbell-shaped mass adjacent to the left side of the T3 vertebra. Magnetic resonance imaging showed a mass extending to the intervertebral foramen at the T3–T4 level and suggested a dumbbell-shaped schwannoma categorized as type 4, according to Eden’s classification. A hybrid spinal endoscopic and robot-assisted surgery was planned for tumor resection. Resection of the tumor within the foramen was first performed using a spinal endoscope in the prone position. Subsequently, the tumor in the thoracic cavity was completely resected by robot-assisted surgery in the right lateral decubitus position. The tumor was removed while preserving the thoracic and intercostal nerves. The excised specimens were pathologically diagnosed as cavernous hemangioma. The patient was discharged after postoperative seven days without any complications or neurological symptoms. Conclusions: Spinal endoscopy is a minimally invasive procedure with limited muscle and bone destruction. For dumbbell-shaped tumors, a hybrid of spinal endoscopy and robot-assisted thoracoscopic surgery may be less invasive and safer option.